topic Re: Please help with this problem. in SAS Procedures

Please help with this problem.

DowChang — Thu, 12 Jul 2012 19:02:43 GMT

I have two sets of SAS tables in the attached Excel file. I want to integrate observations from table 1 to table 2 one-to-one. What existed SAS statement able to do it?

Message was edited by: Dow Chang

Re: Please help with this problem.

Reeza — Thu, 12 Jul 2012 19:19:05 GMT

SAS Learning Module: Match merging files in SAS

Re: Please help with this problem.

MikeZdeb — Thu, 12 Jul 2012 21:44:48 GMT

Hi ... the two tables have the same variable names and different numbers of observations (881 and 896). When you say integrate one-to-one, just what do you mean?

Re: Please help with this problem.

DowChang — Thu, 12 Jul 2012 21:52:50 GMT

Thank you, Reeza

Yes, MERGE and By statements can do one-to-one merging as the SAS Learning Module shown.

That is Match-Merging.

But, my problem is the Non-Match-Merging.

I would like to see the merging result of two tables in the attached Excel file, if you have it.

You may need an algorithm in programming.

Or you have better ideas?

Please correct me if I am wrong.

Dow

Re: Please help with this problem.

DowChang — Thu, 12 Jul 2012 21:56:30 GMT

Please, see my reply to Reeza.

Thanks, MikeZdeb

Dow

Re: Please help with this problem.

Reeza — Thu, 12 Jul 2012 22:36:29 GMT

You've provided the input, but you need to provide an example of what the output would look like.

You can also merge without specifying a by statement, If you just want the two tables side by side, merge without a by but then you'll need to rename all the columns in one of the tables because they have the same names.

data class;

merge sashelp.class sashelp.class( keep=name age sex rename = (name=name2 age=age2 sex=sex2));

run;

Re: Please help with this problem.

DowChang — Fri, 13 Jul 2012 01:16:23 GMT

Reeza,

I provide you with only a part of the output. You can find them in the two tables.

As the example shown, the purpose of this merge is to link observations one-to-one for a piece of information; Obs, in table 2 only.

State	County	Month	Day	Sex	Age	State_D	County_D	Month_D	Day_D	Sex_D	Age_D	Obs
45	45	4	4	M	14	45	45	4	4	M	17	xxxxxx
45	45	4	4	M	29	45	45	4	4	M	28	xxxxxx
45	45	4	14	M	19	45	45	4	14	M	18	xxxxxx
45	45	4	14	M	27	45	45	4	14	M	26	xxxxxx
45	45	7	1	M	35	45	45	7	1	M	34	xxxxxx
45	45	7	1	M	39	45	45	7	1	M	38	xxxxxx

The total output can be produced, but not in SAS at this time.

I would like to know how to produce in the SAS.

Thanks for your comments,

Dow

Re: Please help with this problem.

Reeza — Fri, 13 Jul 2012 01:57:57 GMT

The code example I showed would work in your case then, you just need to change the rename parts

data want;

merge have have (rename= (state=state_d county=count_d etc...);

run;

Re: Please help with this problem.

DowChang — Sat, 14 Jul 2012 03:14:12 GMT

Reeza,

Your help is highly appreciated.

I wish the code that you offered able to get the accurate output.

The only way to find out is to run your code and get outputs.

I cannot run SAS codes now.

I already have the output for the attached two new tables, yet it is not produced in SAS.

I would like to compare other outputs to my output.

Thanks for anyone who may concern.

Dow